View metadata, citation and similar papers at core.ac.uk brought to you by CORE provided by Aston Publications Explorer The hate that dare not speak its name? Robbie Love and Paul Baker This paper uses corpus-based methods to explore how British Parliamentary arguments against LGBT equality have changed in response to decreasing social acceptability of discriminatory language against minority groups. A comparison of the language of opposition to the equalisation of the age of consent for anal sex (1998–2000) is made to the oppositional language in debates to allow same- sex marriage (2013). Keyword, collocation and concordance analyses were used to identify differences in overall argumentation strategies, assessing the extent to which previously explicit homophobic speech (e.g. homosexuality as unnatu- ral) has been replaced by more indirect strategies (e.g. less use of personalised argumentation via the pronoun I). We argue that while homophobic language appears to be on the decrease in such contexts, there is a mismatch between words and acts, requiring analysts to acknowledge the presence of more subtle indications of homophobic discourse in the future. Keywords: homophobia, parliament, debate, corpus, keywords 1. Introduction In 1894, Oscar Wilde’s lover Lord Alfred “Boysey” Douglas published the poem “Two Loves”, which referred to homosexuality as “the love that dare not speak its name”. The poem was mentioned at one of Wilde’s trials for “gross indecency” af- ter which he was imprisoned for two years. Along with Wilde’s incarceration, the poem indicates how during Victorian times, homosexuality in Britain was seen as a criminal offence. Such thinking continued for much of the twentieth century, with medical discourses viewing it as a sickness, newspapers linking it to shame, scandal, deviancy, paedophilia and communism, while religious discourse widely held it to be a sin. However, gradually at first, people began to speak up on behalf of the love that dare not speak its name: homosexuality was decriminalised to an extent in 1967 and despite a backlash in 1988 which forbade its “promotion” by Journal of Language Aggression and Conflict 3:1 (2015), 57–86. doi 10.1075/jlac.3.1.03lov issn 2213–1272 / e-issn 2213–1280 © John Benjamins Publishing Company This is an open access article under a CC BY 4.0 license. 58 Robbie Love and Paul Baker education authorities, since the 21st century much of the earlier discriminatory legalisation has been overturned. This paper is concerned with the language around the legal processes involved in two (successful) attempts to award equality to gay men and women in the UK. The first took place in a series of political debates between 1998–2000 which re- sulted in equalising the age of consent for sexual intercourse for gay men at 16 (it had previously been 18 while the age of consent for heterosexual people was 16). The second set of debates occurred in 2013 and involved allowing same-sex partnerships to be legally recognised as marriages (an earlier Bill had defined such relationships as civil partnerships since 2005, but this was felt to be a compromise by some people). The sets of debates occurred in both the lower (House of Commons) and up- per (House of Lords) chambers of the British government, and had several read- ings each. In the case of the age of consent (AOC) debate, the Lords rejected the proposed Bill three times, causing the lower house to use the Parliament Act to pass the Bill.1 The same-sex marriage (SSM) Bill was twice rejected by the Lords but passed on its third reading. These two sets of debates are a rich source of data for the analysis of discourse and argumentation around homosexuality and equality. While public attitudes have become more liberal towards homosexuality,2 in both debates a substantial number of Members of Parliament and Lords voted against equality, being willing to go “on-record” about their decision, and sometimes speaking at length about why they wished to do so. Considering the shift in public opinion it is pertinent to consider whether and how anti-equality speakers differed over the two time periods in the ways they constructed their anti-equality arguments and attendant representations of gay people. This paper examines the extent and ways in which such arguments and rep- resentations differ between the two sets of debates. We aim to assess the extent to which previously explicit homophobic speech (e.g. constructions of homosexual- ity as unnatural) has undergone replacement with implicit or indirect homophobia (i.e. by accessing discourses that are harder to interpret as homophobic on the surface level, but, nonetheless, very clear in their opposition to LGBT equality). Just as Mills (1998, 247–8) points to more subtle and indirect manifestations of sexist discourse, we aim to show how homophobic discourse “responds” to pres- 1. Since 1949 only four Acts have been passed by the government without consent of the Lords. 2. In 1983 62% of people believed sexual relations between two adults of the same sex was al- most always or mostly wrong, while this figure was 28% in 2012 (British Social Attitudes Survey 2013 Edition Report). The hate that dare not speak its name? 59 sure by transforming its nature. Such discourse can be more difficult to identify and challenge, however, requiring a more in-depth and critical form of analysis. We use a range of approaches from corpus linguistics in order to examine and compare the anti-equality speech from the two sets of debates. Corpus approaches are well-placed to handle large amounts of data. As discussed in Baker (2006, 10– 7) such approaches can reduce researcher bias, act as a form of triangulation and aid the identification of minority positions. We have combined corpus-driven and corpus-based methods (Tognini-Bonelli 2001), the former involving using statisti- cal tests to identify words with comparatively high frequencies in texts, while the latter involves the analysis of a pre-selected set of terms that are felt to be relevant “sites” for discussion of argumentation and representation (in this case words di- rectly relating to homosexual identity and homophobia). After positioning this research in relation to other relevant studies, we discuss how we built and analysed our corpora. This is followed by four results sections af- ter which we conclude with a section which summarises and reflects on the study. 2. Literature review In this section we focus on a small number of key studies that have examined negative representations of gay people via the use of language. Many studies have identified how gay identities are problematized: for example, Kitzinger (2005) used techniques from Conversation Analysis to show how people oriented to references to sexual orientation (either heterosexual or gay) during spoken interactions. In the transcripts she examined, heterosexuality was a taken-for-granted disclosure and did not result in any marked orientation from listeners. On the other hand, during one conversation disclosure of a gay identity became the focus of the con- versation while in another it acted as a conversation killer. The genre of newspaper discourse is perhaps one of the largest areas of anal- ysis of homophobia. For example, Henley et al. (2002) in their study of stories about violent attacks found that the Washington Post used fewer, and less specific nominals when referring to anti-gay violence than to violence against heterosex- ual people. Chirrey (2003) studied a series of newspaper articles regarding the public “coming out” of a pop star, noting how a liberal newspaper used the verb disclosed to frame the act in a relatively neutral non-judgemental way, while tab- loids used more sensational language such as admits, frank admission, secret and in hiding. Morrish (2002) has noted how broadsheet newspapers have used coded references to negatively represent gay men, indicating that coverage of the gov- ernment minister Peter Mandelson included references to homosexual acts e.g. “What Peter Mandelson did was the political equivalent of bare-backing”, while he 60 Robbie Love and Paul Baker was described as camp, hedonistic and narcissistic. Further to that, Baker’s (2005) corpus-based study of two newspapers indicated a number of frequently cited discourse prosodies around the words gay and homosexual, including representa- tions of homosexuality as a behaviour rather than an identity, gay relationships as transient, and gay men as promiscuous, involved in crime, shameless or shameful, politically militant, and proselytising children. Baker (2005) also used corpus-driven techniques to examine the Age of Consent debate, focussing only on the House of Lords but comparing the argu- mentation of those who were for vs. those who were against equalisation. Using a procedure called a keyword analysis (described in the following section) he found that those who argued for equalisation were more likely to construct gay people in terms of their identity rather than their behaviour, and argued that it was wrong to criminalise 16 and 17 year olds for having consensual sex. Also (perhaps stra- tegically), they made reference to the fact that the UK would be forced to pass legislation eventually due to the European Convention of Human Rights. Those who argued against equalisation made more use of historic legal terminology like gross indecency as well as indicating concern for the health and reputations of boys while claiming that the Bill was not needed because the age of consent for anal sex was already equal for boys and girls at 18. They also expressed concern that equal- ity would be the “thin end of the wedge”, leading to demands for further changes to the law (a point which is discussed in more detail in our analysis).